Robot Controller and Manipulator

by

Zulkifli Bin Md Zin

Dissertation submitted in partial fulfilment of

the requirements for the

Bachelor of Engineering (Hons)

(Electrical and Electronics Engineering)

JUNE 2004

Universiti Teknologi PETRONASBandar Seri Iskandar

31750 Tronoh

Perak Darul Ridzuan

roved by,

CERTIFICATION OF APPROVAL

Robot Controller and Manipulator

by

ZuIkiflibinMdZin

A project dissertation submitted to the

Electrical and Electronics Engineering Program

Universiti Teknologi PETRONAS

in partial nilfilmentof the requirement for the

BACHELOR OF ENGINEERING (Hons)

(ELECTRICAL AND ELECTRONICS ENGINEERING)

d. Noh bin Karsiti)

UNIVERSITI TEKNOLOGI PETRONAS

TOONOH,PERAK

June 2004

CERTIFICATION OF ORIGINALITY

This is to certify that I am responsible for the work submitted in this project, that the

original work is my own except as specified in the references and acknowledgements,

and that the original work contained herein have not been undertaken or done by

unspecified sources or persons.

DZIN

ABSTRACT

This project, entitled Robot Controller and Manipulator comprises of development of

a manipulator and controller. The project scopes are to develop an articulated

manipulator with three degrees of freedom (3-DOF), a PC-based robot controller,

capability of robot to determine path plan and some graphical user interfaces. The

invention of robot manipulator aggressively carried out during the industrial

revolution, starting from twentieth century. The robot manipulator is extensively used

in industry since it is believed that its implementation has several advantages over

conventional ways, such as several jobs executed by human are very dangerous,

designed components are sometimes having defects and long period of time taken to

execute repetitive jobs.

Generally, the scope of this project is to develop 3-DOF robot manipulator by

considering dynamic properties. 3-DOF is purposely for positioning function. If the

manipulator is needed to be equipped with manipulating function, it is required higher

than 3-DOF system, usually 6-DOF.

The project is divided into two parts; manipulator design and controller design. The

manipulator design involves construction of manipulator from suitable material,

implementation of forward and reverse kinematics, workspace determination,

movement analysis and finally integration of the designed components.

The controller design consists of development of dc motor drive system, controller

chip by using PIC microcontroller, motor feedback system and graphical user

interface at host PC. The user will control the manipulator through this user interface

and this host PC will communicate with PIC microcontroller through serial

communication.

ACKNOWLEDGEMENT

First of all, I would like to express my highest gratitude to Allah S.W.T for His bless

and help that finally, I able to complete my project.

In preparing this project and report, I would like to acknowledge the important

contribution from the following persons:

My father and mother, for their supportive help,

My supportive supervisor, Assoc. Prof. Dr. Mohd. Noh bin Karsiti,

Supportive lecturer, Mr Haris bin Md. Khir,

Lab technicians, Mr. Isnani bin Alias, Miss Siti Hawa and Mr. Azhar, and

All of my supportive collegues for their help and support.

Also this acknowledgement goes to the companies that involve in my project, where I

get all of my project materials and support:

Eng Huat Hardware Trading,

Perniagaan Aluminium Parit,

Keang Sen Kun Kee Foundry,

IP Bearing & Belts Sdn Bhd,

Syarikat Ipoh Hardware Sdn Bhd,

State Electronic Trading Co, and

Caltrus (Electronic) Sdn Bhd

TABLE OF CONTENTS

ABSTRACT i

ACKNOWLEDGEMENT ii

TABLE OF CONTENTS iii

LIST OF FIGURES v

LIST OF TABLES vii

ABBREVIATIONS AND NOMENCLATURES viii

CHAPTER 1 INTRODUCTION 1

1. BACKGROUND OF STUDY 1

3. OBJECTIVE 2

4. SCOPE OF STUDY 3

5. PROJECT RELEVANCY 3

6. PROJECT IMPLEMENTATION PLAN 3

CHAPTER 2 LITERATURE REVIEW 5

1. THEORY 5

2. CONCEPT DESIGN 7

CHAPTER 3 MANIPULATOR DESIGN 11

1. SELECTION OF ROBOT COORDINATE FRAME 11

2. STUDY ON EXISTING DESIGN 11

3. CONCEPTUAL DESIGN AND DIMENSIONING 12

4. WORKSPACE DETERMINATION 14

5. FORWARD AND INVERSE KINEMATICS 15

6. SELECTION OF GEAR RATIO 17

7. MATERIAL SELECTION 19

8. FABRICATION AND DEVELOPMENT OF MANIPULATOR 20

9. CHANGE OF MANIPULATOR DESIGN 21

CHAPTER 4 CONTROLLER DESIGN 24

1. PATH PLANNING 24

2. RS DC MOTOR 26

m

3. H-BRIDGE CIRCUIT 28

4. OPTICAL ENCODER 31

5. SELECTION OF MCU 33

6. PICmicro 16F877 33

7. DC MOTOR APPLICATION 36

8. THE PROGRAMMING OF THE CONTROLLER 38

8.1 Control Selection 39

8.2 Limit Right and Limit Left Function 39

8.3 Reset Function 41

8.4 Positioning Function 42

8.5 Emergency Function 43

9. DC MOTOR APPLICATION SOURCE CODE 44

10. USER INTERFACE 45

CHAPTER 5 DISCUSSION 46

CHAPTER 6 CONCLUSION AND RECOMMENDATION 49

1. CONCLUSION 49

2. RECOMMENDATION 50

REFERENCES 51

IV

LIST OF FIGURES

Figure 2.1 Robot Controller and Manipulator Block Diagram. 8

Figure 2.2 The Operation ofRobot Controller and Manipulator 10

Figure 3.1 The Components of Robot Manipulator In Terms of Joints 12

Figure 3.2 The Components of Robot Manipulator In Terms of Links 13

Figure 3.3 The Process of Determining Workspace 14

Figure 3.4 Robot Mechanisms, (a) Closed-Loop (b) Open-Loop 16

Figure 3.5 Fitting The Motor into Gearbox 18

Figure 3.6 The Constructed Manipulator 20

Figure 3.7 The Midway-to-Complete Manipulator; (a) Front View

(b) Side View (c) Top View 22

Figure 4.1 Path Planning for Motor Speed. 25

Figure 4.2 Acceleration Behaviors for DC Motor. 26

Figure 4.3 12V 60 rpm RS DC Motor for Joint 1 27

Figure 4.4 Drawing and Dimension of 12V 60 rpm RS DC Motor for Joint 1 27

Figure 4.5 12V 2000 rpm RS DC Motor for Joint 2 and 3 27

Figure 4.6 Detail Drawing of 12V 2000 rpm RS DC Motor 28

Figure 4.7 Torque - Current and Torque - Speed Characteristic of 12V 2000

rpm RS DC Motor 28

Figure 4.8 Application of H-Bridge for Motor Direction 29

Figure 4.9 H-Bridge Circuit 29

Figure 4.10 Modified H-Bridge Circuit by Adding TIP 132 At the MCU

Output Signal. 30

Figure 4.11 The Final Design of H-Bridge 31

Figure 4.12 Encoder Circuit 32

Figure 4.13 PIC 16F877 Pin Layout 35

Figure 4.14 DC Motor Application Schematic Diagram 37

Figure 4.15 The Control Selection Function Flowchart 40

Figure 4.16 The Limit Switch Function Flowchart 41

Figure 4.17 Reset Function Flowchart 42

Figure 4.18 Positioning Function Flowchart 43

Figure 4.19 Emergency Function 44

Figure 4.20 HyperTerminal Interface, Displaying the Current Position for The

Manipulator: Link 1 is 30.99° While Link 2 is 22.99° 45

VI

LIST OF TABLES

Table 1.1 Project Implementation Plan of Semester 1 and 2. 4

Table 3.1 Comparison and Characteristics ofMaterials between Carbon

Steel, Stainless Steel and Aluminum. 19

Table 4.1 Servo Calculation Bandwidth Comparison 34

Table 4.2 Program Memory Resources for Servomotor Application 34

Table 4.3 Data Memory Resources for Servomotor Application 34

Table 4.4 Key Features of PIC 16F877 36

vn

ABBREVIATIONS AND NOMENCLATURES

A/D - Analog/Digital

DOF - Degrees of Freedom.

EEPROM - Electrically Erasable Programmable Read Only Memory

LED - Light Emitted Diode

ISR - Interrupt Service Routine

MCU - Microcontroller Unit

MOSFET - Metal Oxide Semiconductor Field-Effect Transistor.

PID - Proportional, Integral, Derivative

PWM - Pulse Width Modulation.

SCARA - Selective Compliance Assembly of Robot Arm

SPI - Serial Peripheral Interface

UASART - Universal Asynchronous Receiver Transmitter

vm

CHAPTER 1

INTRODUCTION

1. BACKGROUND OF STUDY

Since the industrial revolution, there are increasing demands to improve quality and

reduce manufacturing cost. Traditionally, the products are manufactured by

craftsmen and the quality of the product totally dependant on the skill of the

craftsmanship.

At the beginning of twentieth century (1905), Ford Motor Company introduced the

concept of mass production [1]. This mass production has been proven to reduce

production cost and a big amount of products can be manufactured. This mass

production technique uses production machines and called as hard automation. From

that, the process automation widespread until the introduction of robot manipulators

that able to perform specific tasks, such as material handling, spot welding, spray

painting and assembling.

In this project, a robot controller and manipulator need to be designed. The main

characteristics of this robot controller and manipulator are movement in three degree

of freedom (3-DOF) and manipulator design with kinematics properties. The degree

of freedom is the number of independent inputs required to precisely position all links

of the mechanism with respect to the ground. The number of DOF of a mechanism is

also called the mobility. In this project, by specifying three-coordinates system (x, y

and z) is adequate to represent any points in space.

The aim of this project is to develop an articulated robot manipulator and controller

that exactly locate at any position in space with 3-DOF. To the extent of application

of this design, the robot manipulator can be improved by introducing machine loading

application, pick and place operation, welding, painting, inspection, medical

application and remote location.

2. PROBLEM STATEMENT

At the beginning of the project, several problems have been identified that yield the

relevancy of implementing this project as listed below:

• Humans are sometimes exposed to the hazardous environment at workplace.

• Forming and joining of designed components usually problematic especially

in automotive industry

• Longer period of time taken for human to execute repetitive works.

• Technical fault may occur as the result of careless and inadequate skills of

worker.

• It is dangerous and not worth for human to carry out hazardous work

especially in military, underwater and remote locations.

3. OBJECTIVE

The objectives of this project are as follow:

• To carry out research, design, implement and fabricate an articulated 3-DOF

robot controller and manipulator with the ability to locate the manipulator

exactly at any position in space.

• To design a robot controller and manipulator with specific kinematics

properties.

4. SCOPE OF STUDY

The scope of study could be understood as per described below:

• Manipulator design with the selection of appropriate material especially that

having light property. The manipulator is composed of three joints and those

joints are designed in such a way that having minimum friction and reliable to

move in its workspace. The gearing and belting system also specified under

manipulator design.

• Controller design hearted by four PIC microcontrollers that have main

functions to drive the motor appropriately and communicate with PC.

5. PROJECT RELEVANCY

The demand for robotic manipulator, especially in fully automated manufacturing

company keeps increasing for at least two reasons; firstly, to acquire fast processing

time and secondly, to avoid manufacturing defects commonly done by human. In

factory, most of the processes are done by automated machines and human will only

monitor that process or work at the end of production line.

As the technology keeps changing, the robot manipulator has been experiencing

advancement of its design and construction to cater the industry's needs. Thus, the

study and designof robotcontroller and manipulator could be seen as relevantproject,

as to understand and follow the technology in robot manipulator.

6. PROJECT IMPLEMENTATION PLAN

The project planning is divided into two frames of time, Semester 1 and Semester 2.

Table 1.1 shows the project implementation plan for both Semester 1 and Semester 2.

Table 1.1 Project Implementation Plan of Semester 1 and 2.

Division Semester 1 Semester 2

Manipulator Design • Study of existingdesign.

• Manipulator design(drawing anddimensioning).

• Workspacedetermination.

• Forward and reverse

kinematics.

• Gear design.• Finding the most

suitable material for

robot manipulator.• Fabrication of robot

manipulator.

• Touch-up on thefabricated

manipulator.• Integration of

controller and

manipulator.

Controller Design • Path planning.• Pittman motor testing

and understanding ofits operation.

• Development of H-Bridge circuit.

• Microcontroller

selection.

• Microcontroller

programming.• Development of

software, as aninterface between PC

and robot controller.

• Development of motordrive system.

CHAPTER 2

LITERATURE REVIEW

1. THEORY

The word robot means different thing to different people. Robot is defined as a

mechanical device that is capable of performing human tasks or behaving in a human

like manner [2], Robot also means a device that controlled by computer and easily

reprogrammable. Thus, manual handling devices (i.e., a device that have multiple

degrees of freedom and it is actuated by an operator) or fixed-sequence robots (i.e.,

any device controlled by hard stops to control actuator motions on a fixed sequence

and difficult to change) are not considered to be robots [3]. Another definition is a

robot is a re-programmable multi-functional manipulator designed to move material,

parts, tools or specialized devices, through variable programmed motions for the

performance of a variety of tasks [4].

Based on definitions above, we define robot as a combination of mechanical and

electronic device that can be multi-functionally reprogrammable to execute a specific

task from variety of tasks depend on its program.

The word manipulator means the executor of process of handling or controlling

something with skill or predefined program that is made up of several links connected

by joints. The word controller means a device that has data processing ability to

instruct, manage and direct specific function or task from data inputted.

When talking about robot manipulator, there are several important elements that

related to manipulator, which are links and joints, robotic system and robot

classifications. The individual bodies that make up a mechanism are called links.

The connection between two links is called a joint. The number of degrees of

freedom (DOF) depends on the number of links and joints and the type of joints used.

DOF is the number of independent parameters or inputs needed to specify the

configuration of mechanism completely. There are five basic types of joints that

frequently used in robot manipulator, which are revolute joint, prismatic joint,

cylindrical joint, helical joint and spherical joint. In this project, revolute joint type is

selected which it permits two paired elements to rotate with respect to each other

about an axis.

Robotic system consists of three major parts, which are mechanical manipulator, end

effectors and controller. Manipulator and controller have been defined above, and the

end effectors are devices that connected to the output link of a mechanical

manipulator to grasp, lift and manipulate works piece. In this project, the design only

considers the development of manipulator and controller without end effectors since

the required DOF is three. In most cases, the first 3-DOF is important for positioning

and normally another 3-DOF is required for manipulating. Thus, to obtain a complete

manipulator, it requires at least 6-DOF, 3-DOF is quite similar with the representation

of a point in space with three coordinates system as shown in the equation below.

P=aJ+byj +c,k (1)

Normally, the robot is classified by DOF, kinematics structure, drive technology,

workspace geometry and motion characteristics. The kinematics structure is

differentiated by open-loop and closed-loop chain, drive technology by electric,

hydraulic or pneumatic, workspace geometry by cylindrical robot, articulated robot or

spherical robot and finally motion characteristics by planar or spherical mechanism.

In this project, an articulated, electrical drive 3-DOF of robot controller and

manipulator will be developed. Since the articulated type is chosen, the joint is

selected as revolute type. Under articulated type, all joints must revolute.

The designing of robot manipulator deals with kinematics that specify the aspects of

motion without regard to the forces or torques that cause it. The kinematics deals

with the positioning of manipulator, velocity and acceleration of the position variables

with respect to the time. The joint variables are related to the positioning of the

manipulator and kinematics study is said as focal point of robot manipulator since it

relates positioning of manipulator with the acceleration, velocity and position

variables. That is why the kinematics study is the first concern in robot manipulator

design.

However, kinematics study has two different approaches; kinematics analysis and

kinematics synthesis. Kinematics analysis deals with the derivation of relative

motions among various links of a given manipulator [5]. There are two types of

kinematics analysis, which are forward kinematics and inverse kinematics. For the

case of to find the set of position and orientation, the corresponding time derivative

which will bring the manipulator to the desired position is called inverse kinematics.

On the other hand, the joint variables and its time derivative is obtained from sensor

or reading, and this data is used to position and orientate the manipulator is called

forward kinematics.

Kinematics synthesis is the inverse process of kinematics analysis. In this case, with

certain desired kinematics characteristics, the manipulator can be developed. For

example, with a given time derivative and position of manipulator, corresponding

joint variables and the type of geometry and manipulator can be determined.

2. CONCEPT DESIGN

In general, robot manipulator consists of two parts; mechanical and electronics. The

mechanical part usually related to physical design and construction of manipulator

whereas electronics part is related to controller and drive system. Figure 2.1 shows

the general block diagram for controller and robot manipulator.

The function of microcontroller is to receive instruction commanded by user, converts

the input into digital signal and finally sources the signal to the h-bridge to rotate the

motor. The rotation of the motor is indicated by high and low signal supplied by

microcontroller into two h-bridge inputs. The coordination of 10 or 01 at a time will

result the motor to rotate either clockwise or counter-clockwise while 00 will give no

signal, thus the motor will static. The microcontroller used is PIC 16F877,

manufactured by Microchip. The details about selection of microcontroller and the

concept of dc motor application by using PIC 16F877 is discussed in the Chapter 4

under Controller Design.

The motor is specified to rotate in its workspace with the specific degree of rotation.

The RS motors with add-on encoder are chosen to be used in this design. Once the

motor is rotating, encoder circuit will detect the rotation of motor's shaft and

generates a pulse signal. A pulse signal generated is very significant to the degree or

distance of rotation, and inputted back to the microcontroller. In order to achieve the

required degree of rotation, microcontroller will do some calculation works and will

stop the motor as the pulse count achieved.

—•

—•

—p.

H-bridge Motor &Gearbox

Manipulator1st joint

PC H-bridgeMotor &

Gearbox

Manipulator2nd jointU-controller

H-bridgeMotor &

Gearbox

Manipulator3rd joint

Figure 2.1 Robot Controller and Manipulator Block Diagram.

The motor is not directly driven from output pin of microcontroller, but using a

device, called h-bridge. This circuit will act to determine the direction of motor,

normally set as clockwise or anti-clockwise and to totally stop the motor. Note that,

this circuit cannot control the speed of the motor and only microcontroller is able to

control the speed with the variety of input voltage signal supplied to the motor.

The function of motor is to move or revolute the joint. One motor is assigned for a

joint. The rotational angle of motor must be accurate to ensure that manipulator is

exactly located at desired coordinate. Manipulator first joint, second joint and third

joint have the task to position at any coordinate in 3-dimension workspace,

particularly in x-axis, y-axis and z-axis.

In this design, a PIC 16F877 microcontroller is used. The controller controls a motor

and responsible for positioning the joints in its axis. This microcontroller also

programmed to communicate with PC. The communication with PC is carried out by

using RS-232 connection through serial communication.

In order for controller to communicate with PC, the interface must be developed

where software in PC and programming for microcontroller must be established. This

user interface is developed by using Visual Basic or HyperTerminal.

The relationship between computer software and microcontroller is like this; the user

will specify the location or coordinate for manipulator in computer program,

computer program will convert it to instruction that understood by microcontroller

and finally that microcontroller will control motor's operation. The motor will rotate

accordingly and stops if the required degree is achieved indicated by adequate

encoder pulse signal. Figure 2.2 shows how robot manipulator and controller work.

INPUT

The command is invokedthrough PC to PICmicrocontroller by usinga user friendly interface.

PDIP

The output fromencoder isreturned back to

PIC to send

signal to h-bridge to stopthe motor.

Finally, after the PIC sendssignal to stop the motor, thePIC also send together thelatest state/position ofmanipulator. The position willbe updated by interface foruser reference.

OUTPUT

To output from PIC isconnected to h-bridge todrive the motor.

Figure 2.2 The Operation ofRobot Controller and Manipulator.

10

CHAPTER 3

MANIPULATOR DESIGN

1. SELECTION OF ROBOT COORDINATE FRAME

As has been discussed earlier, the manipulator has several coordinates, namely

cartesian, cylindrical, spherical, articulated and Selective Compliance Assembly

Robot Arm (SCARA). The illustration for these types of robot coordinate is shown in

Appendix A.

For the purpose of this project, an articulated coordinate frame is more interesting

than the others. The reason is an articulated robot's joints are all revolute and similar

to human arm. This configuration is very famous and perhaps mostly used in

industry. The advantage of this configuration over other types is an articulated has

greater accessibility in its workspace with the same robot's dimension.

2. STUDY ON EXISTING DESIGN

A study on the existing design has been conducted to find the ideas for manipulator

design. The area of interest is only on developing mechanical and physical design

and manipulator's orientation. The observations include the method for eachjoint is

connected, physical orientation and arrangement and shaft connection with

manipulator's joint.

11

Study and observation of design has been made in Industrial Automation Lab and

through robotic books and sites. Among the type of robots that has been studied are

Mitsubishi Robotic Arm, ED Robot 5-Joint Robotic Arm Trainer, Motoman ES165,

Adept-One robot and Fanuc S-900W robot. The illustration of these robotic types is

shown in Appendix B.

Among the important observations found is the arrangement of motor, which to make

sure that the accessibility of manipulator at its maximum with the proper arrangement.

Another observation is the mounting of gear and connection ofjoint with the base and

joint with the joint.

3. CONCEPTUAL DESIGN AND DIMENSIONING

The design ofrobot manipulator consists of four parts: a base,joint 1,joint 2 andjoint

3. Figure 3.1 and 3.2 shows the component ofrobot manipulator.

Figure 3.1 The Components of Robot Manipulator In Terms of Joints

12

Figure 3.2 The Components ofRobot Manipulator In Terms of Links

Controller box and dc motor for joint 1 are placed in the base. Joint 1 is used to rotate

the manipulator with respect to z-axis, joint 2 is used for y-axis and joint 3 is also for

y-axis. Even though there is no specific joint for rotating with respect to x-axis, by

catering the rotation in both y- and z-axis for all joints, any 3-dimension position in its

workspace can be achieved. With the accurate revolution of joint 1Joint 2 andjoint 3

ensures that the exact location of manipulator could be established. Appendix C

shows the revolution of eachjoint with respect to the specific axis, indicated by 9i for

first joint's revolution angle, 62 for second joint's revolution angle and O3 for third

joint's revolution angle.

The base width is 11.00 inches. The wide base compared to joint's width is important

to make sure the stability of manipulator is optimum. The height for base, link 1 and

link 2 are 6.9 inches, 9.3 inches and 7.00 inches respectively. The height of this

manipulator, as the all joints are fully stretched upward achieves 26.275 inches

13

including the space occupied by base gear and base joining. The illustration for

manipulator dimension and its front, side and top view could be seen in Appendix D.

The manipulator design uses two units of adaptable gearbox, to suit with two dc

motors. The function of these gearboxes is purposely for speed reduction which robot

manipulator normally requires quite slow movement, approximately 20 rpm.

4. WORKSPACE DETERMINATION

Workspace is defined as the volume of space the end effectors can reach [6]. Two

different types of workspace are frequently used which are reachable workspace and

dexterous workspace. A reachable workspace is the volume in workspace which

every point that can be reached by end effecters by at least one orientation while

dexterous workspace is the volume in workspace which every point can be achieved

with any orientations.

Figure 3.3 The Processof Determining Workspace

14

The workspace of manipulator is firstly determined by specifying the limit of

rotational angle for eachjoint. In the earlier design stage, the limit of rotational angle

is determined and then the volume of workspace that can be achieved by manipulator

might be developed. Then, the movement of robot's manipulator is analyzed to

determine its workspace. Appendix E shows the workspace for the manipulator.

Thus, from the design, the workspace of rotational angle for each axis can be

developed as follow:

-150° < 0,

In a 1-DOF system, all variables can be known upon specification of any variables.

For example, when a variable is set to a certain value, the mechanism is totally set and

other variables are known. Normally, the mechanism consists of closed-loop and

open-loop. Most of the closed-loop system mechanism is in the form of 1-DOF. The

example of closed-loop mechanism is given by closed-loop four bar mechanism,

which if we set the movement of one link to certain degree, let say 120°, the other

links' angle also can be detected.

3-DOF robot manipulator is open-loop mechanism. Unlike in the closed-loop

mechanism, if all of the variables are specified in open-loop system, there is no

guarantee that the manipulator can exactly position at the given location. This is

because if there is any deflection in one link, it may affect the location of the

manipulator. As a result, the construction of 3-DOF robot manipulator must strong

enough to eliminate all deflections. Figure 3.4 below shows the closed-loop and

open-loop mechanisms.

Figure 3.4 Robot Mechanisms, (a) Closed-Loop (b) Open-Loop

Another difference between closed-loop and open-loop mechanism is closed-loop

mechanism provide feedback while open-loop mechanism is does not. In a feedback

system, any deflection of a link or error occurs can be detected. For example, if

deflection occurs at link AB, the link O2B will detect the changes but in open-loop

16

system, there is no detection since the deflection will move all succeeding members

without feedback.

To remedy this problem in open-loop robot, the position of the hand is constantly

measured with devices such as camera, the robot is made into closed-loop system with

external means such as the use of secondary arms or laser beams [7].

In this project, the position analysis is done by analyzing at the given location or

coordinate, what is the angle required for joint 1,joint 2 andjoint 3 to revolute. This

approach is done by applying trigonometric function. For example, if an articulated

3-DOF robot manipulator in this design want to move to location from (0,0,0) to (7,

4,10) the joint variable required is as below:

6> - +29.75°

There are several types of gear, namely spur, helical, bevel and worm. In this project,

two sets of spur gear combined in two units adaptable gearbox, manufactured by RS

are used. Spur gear has teeth parallel to the axis of rotation and used to transmit

motion from one shaft to another parallel shaft.

The motor used in this project is RS 12V dc motor. From data sheet, the nominal no

load speed for this motor is 2000 rpm. The gear ratio calculation is as below.

IfNi is the speedfor motor and N2 is the required speed, the gear ratio needed to

reduce the speedfrom 2000 rpm to 20 rpm will be

Gear ratio =N2

_ 2000rpm20rpm

= 100:1

(4)

It is fortunate since RS provides a gearbox with the ration 100:1. The torque for this

adaptable gearbox is 4 Nm. Thus, with the proper gear ratio and torque, this

adaptable gearbox is used in the design. The adaptable gearbox is assembled together

with RS dc motor. Figure 3.5 below shows the arrangement of RS dc motor and

adaptable gearbox.

D

Figure 3.5 Fitting The Motor into Gearbox

A - Adaptor PlateB - Pinion

C - Motor

D- 12.5 mm

E - Gearbox

The size of gear is expressed as pitch, which is roughly by counting the number of

teeth on the gear and dividing it by the diameter of the gear. In order to ensure that

the transfer of energy between gears is efficient, the pitch between these gears must

be the same. The same pitch also important since different value of pitch may result

easy to wear over several period of operation time and non-matching between gears.

7. MATERIAL SELECTION

There are several suitable materials to be used in fabricating the manipulator. Among

the materials that can easily be found in market are carbon steel, stainless steel and

aluminum. Table 3.1 below shows the comparison and characteristics of these

materials.

Table 3.1 Comparison and Characteristics ofMaterials between Carbon Steel,

Stainless Steel and Aluminum.

Characteristic Carbon steel Stainless steel Aluminum

Weight Highest Medium Lightest

Strong Acceptable Acceptable Acceptable

Cost Cheapest Medium Highest

Easiness to fabricate (in

terms of easiness to drill, cut

and shape)

Lowest Medium Highest

Thus, from the above comparison, aluminum is selected as material for constructing

robot manipulator. The highest weightage of material characteristic falls to materials'

weight since we need to find the lightest material as possible in order for motor to

revolute thejoint efficiently. Note that, forRS motor, the maximum affordable torque

is 4 Nm, if equipped with adaptable gearbox.

From the comparison, the price for aluminum is the highest. Even its price is high, it

is still affordable to use this material for fabricating purpose.

19

8. FABRICATION AND DEVELOPMENT OF MANIPULATOR

The fabrication of manipulator's model is carried out in the lab. The fabrication

process needs machining and drilling skills, and all other skills that related to metalwork. The process of constructing robot manipulator takes quite long time and the

most important thing during the fabrication process is the machined and cut materialis capable to combine together with a stable joint to develop several strong links.

Figure 3.6 below shows the picture ofdeveloped manipulator.

Figure 3.6 The Constructed Manipulator

20

It is targeted that, the manipulator is constructed from used material as to minimize

the total cost, since the cost for new material, especially aluminum is quite high. The

suitable used materials can be obtained from thorough finding process from used

equipments. Among the potential places to get this such materials are second hand

goods dealer, bicycle, motorcycle and car workshop and used equipments at home e.g

VCRs, CD player, washing machine and mechanical toys. In case of there is difficult

to find used material for certain manipulator's part, new material is chosen.

In manipulator design, we have specified our own dimension and all materials needed

must be suited with the dimension. The most probable problems for this condition is

unable to find the suitable specific material for needed parts. That is the disturbance

and defector while doing robotic project. If the material cannot be found over set time

range, there is a solution here which is by fabricating ourselves the manipulator's

parts.

9. CHANGE OF MANIPULATOR DESIGN

The current manipulator design has been reviewed for several times. The initial

design is quite distinct compared to final design. As noted, the big differences

between initial design and final design particularly on its gear implementation and

type of motor used.

During initial design, the design involves a lot of gear to be located at each link, to

rotate the joint. For example, instead of gear boxes used, there is another gear

purposely to mount the joint with motor, driven by belt. Several problems have been

identified and difficult to solve, which are difficult to find the suitable gears with

specific gear ratio and belt that suits with that gear. Most of the industrial suppliers

only supply belts or gears for heavy usage for example big machines, tractors and

other industrial equipments. The detail of initial robot manipulator design is shown in

Appendix G. The pictures for initial design are shown in Figure 3.7.

21

Figure 3.7 The Midway-to-Complete Manipulator; (a) Front View (b) Side View (c)Top View.

Thus, due to unsolved problems for nearly one semester, achange of design must becarried out This new design (Appendix D) implements as less as possible the usageof gear. It stresses more on implementation of adaptable gearbox and direct drivefrom the gearbox to manipulator's joint. According to this design, the motor used forinitial design which Pittman motor need to be changed to RS since RS motor offers an

22

attractive adaptable gearbox. Furthermore, the gearboxes offered by RS are having a

lot of gear ratio varieties. In the new design, several suitable couplers also must be

fabricated for the purpose of to obtain direct drive from the output of gearbox to

respectivejoint.

23

CHAPTER 4

CONTROLLER DESIGN

1. PATH PLANNING

In the previous chapter, under forward and inverse kinematics topic, the kinematics

analysis of robot manipulator and its links has been discussed. Under kinematics

analysis, by usingequationof motion, we can knowwhere the positionof manipulator

will be if we have joint variables. Path planning actually refers to the way of robot

will move from one location to another location. In manipulator study, the movement

ofrobot will be in two types: the normal movement and relative movement.

Normal robot movement is specified as the movement of manipulator to certain

position from origin. For example, the positioning of manipulator from (0, 0, 0) or

base to (4, 10, 3) is said as normal movement. On the other hand, the relative

movement is specified as the movement of manipulator from specificcoordinate point

to another coordinate point, except base coordinate. For example, the movement of

manipulator from coordinate point (4, 10, 3) to (-3, 9, 5) is said as relative movement

since the positioning is referred to as another specific point, not at the origin. During

setting up the equation of motion, the movement of manipulator always refers to the

difference between these points.

A path is defined as a sequence or robot configurations in a particular order without

regard to the timing of these configurations. So, if a robot goes from point (and thus,

configuration) A to point B to point C, the sequence of configuration between point A

and B and C constitutes a path [8]. The most important concern in developing a path

24

is the movement of robot from one point to another point regardless of the time

consumed to achieve the desired point.

As the concept applies, in order to move a robot manipulator from a point to another

point, the microcontroller signals motor to rotate the gear together with the arm. By

realizing the motor behavior and manipulator's path planning, it will accelerate at the

beginning of operation until motor's speed reaches maximum speed, indicated as

phase 1 and decelerate at the end of operation, indicated as phase 2. For instance, let

say if we want to move a link for 10°. The microcontroller gives a signal to motor

and motor will rotate. The encoder signal pulse becomes another input to the

microcontroller. If an encoder signal for each high pulse indicates 1° of rotation, it

requires 10 pulses for microcontroller to stop the motor rotation.

The motor will exactly locate an arm if the speed of motor always constant, starting

from beginning until the end. But the reality is not that way. In path planning

concept, motor acceleration takes several times, then the motor will be constant in its

speed and finally decelerates for several times, Figure 4.1, yielding a small variance in

the final destination.

Motor speed, a>

Constant speed phase

Accelerationphase

Total time, t

Deceleration

phase

10 pulses Time, s

Figure 4.1 Path Planning for Motor Speed.

Note that, the variance in final destination is due to variance in speed, smaller during

both acceleration and deceleration phase, making the total travel would be lesser. It

will be getting a clear picture from graph of acceleration, as shown in Figure 4.2,

25

which positive during acceleration phase, zero during constant speed and finally

negative for deceleration phase.

J v Motor acceleration, a

Time, s

Figure 4.2 Acceleration Behavior for DC Motor.

The above-mentioned problem is a kind of challenge in path planning. This problem

is difficult to eliminate but can be minimized by using light material for manipulator

and also using intelligent microcontroller that can compensate between operation

time, acceleration and distance traveled. Beside that, error calculation function in

source code is very useful to minimize this effect.

2. RS DC MOTOR

There are three motors used in this robotic project, type RS 12V dc motor. The motor

has ratings of 12V input voltage, 2000 rpm nominal speed at no load and 4 Nm

maximum torque when attach to adaptable gearbox.

This motor has a built in encoder at the bottom part. The function of encoder is to

calculate the rotation generated by the motor, indicated by pulse. A pulse generated

equals to a rotation has been done.

26

4139

Figure 4.3 12V 60 rpm RS DC Motor for Joint 1

t - 4 MOLES2,7 DIA"ft "

32-

34-

ZB-- T"

1-5—1 -37-

[•nin- ,.•d.^-i""™!?^" *

loS?

n

1.5

-27

•'•'toatw

Figure 4.4 Drawing and Dimension of 12V 60 rpm RS DC Motor for Joint 1

Figure 4.5 12V 2000 rpm RS DC Motor for Joint 2and 3

27

47.7

84.8

20M3 2.5 -+-'K

5-H

25.9

-aw' -0.012

on

Figure 4.6 Detail Drawing of 12V 2000 rpm RS DC Motor

mfv

120

80

40

1

.m Vt\bJ,m

48V 24V 12V/120

BO

40

§y^v_ / // / /_^

n _i ' / /

2.-•

-12V •I

/T 7 I

t /[ ^

^fs - I i \'234 0 1ooo 2O0O 3000

rorque/Gurrent Torque/Speed

Figure4.7 Torque- Current and Torque - Speed Characteristic of 12V2000 rpm RS

DC Motor

3. H-BRIDGE CIRCUIT

Another aspect under controller design is development of h-bridge circuit. The

function of h-bridge is to control the direction of motor, whether clockwise or anti

clockwise. This circuit actually does not have the ability to control the motor speed.

In robotic design, it is desirable to change the direction of motor by changing the

direction of current flow. In other words, if the h-bridge is connected to

microcontroller, the changing of motor direction can be performed by changing the bit

information signaled to h-bridge circuit.

A simple h-bridge circuit can accomplish this as shown in Figure 4.8. As the output

from microcontroller change from high to low or vice versa, the direction might be

28

changed. When the transistor SW1 and SW4 are high, the current flows in the

direction from A to B. If the transistor SW2 and SW3 are high, the current flows in

the direction from B to A. Thus, the polarity will change and motor direction also

will be changed.

-i:Y-

- L0W: -j. " lagiK. ••--—&

Figure 4.8 Application ofH-Bridge for Motor Direction.

The circuit inFigure 4.8 is a simple andfor illustration purpose only on its operation.

The h-bridge circuit design in this project is shown inFigure 4.9.

MCU_agnail

D—, •_

kR3"IQkohrr

HCU_Signal

• 1

R2

2.2kchm

.R1

L10hohm

R4'22kofti

!Q2TIP132

Q1TIM 32

D12\lH400d

D2DC MOTOR

7\ 1H4004-

T

D3 G3 l.T1P132 _>

D47^1H40Q4 Q4

TIP132

Figure 4.9 H-bridge Circuit

T

V1

,12V

In h-bridge circuit above, Rl and R3 consist of 10 Ul pull-up resistors. Its function isto eliminate jerk when the microcontroller powers up and down. There are four

D1N4004 diodes used. Dl and D3 function is to protect the chips from over voltage

29

by turning on when more voltage is coming from the motor than is coming from the

batteries. Finally, D2 and D4 protect the chips from under voltage by turning on

when the voltage in the motor is below GND.

The h-bridge circuit shown in Figure 4.9 is the first design of motor drive control. It

uses low complementary silicon power darlington transistor, TIP 132 that has current

gain, hFE typically at 1000. The required current to drive dc motor is 0.5 A at its

nominal speed. Even the current gain is quite high for TIP 132,the transistor still not

able to provide sufficient current to drive the dc motor. Note that, the measured

output current from PIC output pin is 0.02 mA.

As the above problem is identified, the design is modified by adding TIP 132 at the

PIC output pin to amplify the MCU signal as shown in Figure 4.10. The result is also

the same where the motor cannot operate properly.

MCU_Slg*a11

p

MCU_Si^ial2

•

TIP132

Q5

TIP132

Q6

,H3MOkohir

R2

22Vo\\m-R1»10kohm

02TIP132

<-R4-22kohm

Q1TIP132

D1A 1M4O04

DZ

1W004£'"~ £1

^>

T

, D3

MCU_Signd1

•

MCU_$ignd2

•

M3

BS170

m

BS17Q

TIP132

Q5

TIP132

Q6

j.R3MOkohm >Ms >22kdim

R2

-wv-

TIP132

Q1TIP 132

D1

^\ 1NdflOd

D2

AlN4Mil

^>

D3 Q3kl£l)WW Tlf>132 J

D4 04AlW0WTHtt

VI,12V

V2

_8V

Figure 4.11 The Final Design of H-Bridge

For the implementation of this project, a specific h-bridge IC, L6203 manufactured by

SGS-Thomson is also used. This kind of driver IC uses a technology called

Multipower BCD which allows the integration on the same chip of isolated power

DMOS elements, bipolar transistors and C-MOS logic. The maximum supply voltage

that this IC can withstand is 60V, maximum current of 1.5A and total power

dissipation of 1.5 W.

4. OPTICAL ENCODER

Optical encoder is one of the famous encoders since it is least expensive to buy and

the easiest to build its parts. The encoder consist hollow disk or marked transparent

diskand an optical sensor. Commercial encoders have segments in the range of 16 to

1024. The more segments we have, the finer the measurement is possible.

There are two ways to detect light and dark pattern on the disk which are transmission

and reflection. In transmission type, the lightwill go through transparent segment and

received by phototransistor or photodiode. On the other hand, in reflection type, the

emitter and receiver will be placed on the same side of disk. The light will be

reflected by white areas while dark areas will absorb the light.

31

In this project, transmission type is chosen. For transmission type disk, the pattern

can be printed on clear plastic or heavy white paper with a laser printer. Then, the

pattern is laminated to ensure its toughness and to avoid scratch in case of

accidentally contact with the detector or environment.

By utilizing an encoder, the idea is this device will give the pulse for every angle of

rotation. A pulse given by an encoder represents certain angle of rotation. Let say

that the encoder disk has 64 segments (32 white and 32 black). The angle of one

pulse is suggested by calculation below:

360°Angle per pulse

32 segments= 11.25° per pulse (5)

The pulses generated by this encoder become an input to microcontroller. For every

process, the microcontroller will remember the last pulse generated by encoder and

will count the new pulse inputted from encoder. If the required angle is achieved, the

microcontroller will send a signal to motor to stop the movement. Figure 4.12 below

shows the circuit for encoder.

V1

9V -=±=

R1

AA/V2.2k

IR LED\L

PHOTO

TRANSISTOR

Na

Q2<

R2

5. SELECTION OF MCU

There are various types of MCU in market in the manufacturing line of Microchip

Technology Inc. First thing to know is there are lot of MCU manufacturers instead of

Microchip Technology Inc such as Motorola, Intel, Atmel, Siemens and several other

companies. The selections of MCU from production line of Microchip Technology

Inc are based on its program or source code that easily understood, the MCUs that

easily obtained and wide availability of its resources in terms of its program, source

code and programmer. The type of MCUs is usually differentiated by their function,

pin arrangement and application; PDIP, SOIP, SSOP, PLCCand TQFP.

The MCUs also are differentiated by their functions or applications such as EEPROM

capacity, AID conversion, serial communication, addressable universal asynchronous

receiver transmitter (UASART) andpulse width modulation (PWM) function.

For controlling robot manipulator function, it requires serial communication function

to communicate between PC and MCU. Furthermore, the function also requires

PWM generated signal to drive the motor through h-bridge. Beside that, since the

program for controller is in the form of receiving previous input, stores it and

compare to the newgenerated input, it requires bigEEPROM capacity.

Thus, based on these applications needed, PIC 18C452 and PIC 16F877 meet the

above features and the selection is based on the availability of that microcontroller in

market and selling price.

6. PICmicro 16F877

There are two types of Microchip microcontrollers that are very suitable for this

project, namely PIC 18C452 and PIC 16F877. The suitability is measured to the

features that owned by microcontrollers.

As discussed earlier, the controller project for robot manipulator requires at least the

following functions; EEPROM capacity, A/D conversion, serial communication,

33

addressable universal asynchronous receiver transmitter (UASART) and pulse width

modulation (PWM) function. The PIC 18C452 and PIC 16F877 are said suitable for

this project since it has the above-mentioned functions. The comparison between PIC

18C452 and PIC 16F877 is shown in specific tables below.

Table 4.1 Servo Calculation Bandwidth Comparison

DeviceOperatingFrequency

Hardware

Multiplier

Servo

UpdatePeriod

Maximum

Servo Calculation

Time

MCU Bandwidth

Used

PIC16CXXX 20 MHz Mo 563 fisec 540 fisec 96%

PIC18CXXX 20MHZ Yes 563 fisec 97 ^isec 17%

Table 4.2 Program Memory Resources for Servomotor Application

DeviceAvailable Program

MemoryProgram Memory Used

by ApplicationPercentage Used

PIC16F877 8192 X 14 3002 X 14 37%

P1C18C452 32768 X 8 8265 X 8 25%

Table 4.3 Data Memory Resources for Servomotor Application

DeviceTotal Data Memory

Available

Data Memory Used

by ApplicationData Memory Used

by Compiler*1'Available Data

Memory

PIC16F877 368 bytes 78 bytes 44 bytes 246 bytes

PIC18C452 1536 bytes 78 bytes 386 bytes 1072 bytes

Note 1: The amount of data memory will depend on the compiler used- This memory is used Tora software stack,temporary variable storage, etc.

The difference bandwidth used between PIC 16CXXX and PIC 18CXXX is quite

significant, showing that PIC 18CXXX has greater bandwidth. From the table above,

there is 17% of bandwidth used for PIC 18CXXX compared to 96% of bandwidth

used by PIC 16CXXX. Again, for program memory resources, PIC 18C452 has

lower percentage usage than PIC 16F877 with 25% and 37% respectively. From the

last table above, it shows that PIC 18C452 has greater data memory resources. Thus,

it is concluded that PIC 18C452 is better than PIC 16F877 in terms of its

performance.

When come to the decision process, which one microcontroller to be chosen, the

requirement might be expanded to certain extent, which another factors might be

included. In this case, there are at least two criteria that are interrelated comprising of

34

availabihty in market and cost. Basedon these points, below are the comparisons of

these microcontrollers. PIC 16F877 is easier to obtain in market compared to PIC

18C452. Furthermore, PIC 16F877 is cheaper than PIC 18C452 by the difference of

price reaches RM 20.00 each. Thus, upto this point, PIC 16F877 is chosen.

The above paragraph discusses in general the requirements needed for the project.

This paragraph, in turn will explain further about PIC 16F877 chip. PIC 16F877 is

CMOS flash-based 8-bit microcontroller. It features 256 bytes of EEPROM data

memory, selfprogramming, anICD, 8 channels ofAnalog-to-Digital (A/D) converter,

2 additional timers, 2 capture/compare/PWM functions, the synchronous serial port

can be configured as either 3-wire Serial Peripheral Interface (SPI) or the 2 wire

Integrated Circuit bus and a Universal Asynchronous Receiver Transmitter (USART).

PIC 16F877 comes together with 40 pin package. The pin layout chosen is PDIP.

The microcontroller layout andpinassignment could be seen as in Figure 4.13 below.

Onthe otherhand, the key features of PIC 16F877 is shown in Table 4.4 compared to

other PIC 16 series.

PDIP

^.CLRj'vpp -

RAtt'AND

RA1?AN1

RA2'AN2iV'.REF-

RA3/AN3/Vre=+

R.M/T0CK1

3vAS/AN4/$S

REQ/RD/AN5

REI/WR/AN8

Rt2/C3/AN7Vcc

VS3

OSCVCIKIN

OSCl'CLKOUT

RCGTIOSO/TIGK

3cimost/ccP2

RC2/CCF1

RC3SCK/3CL

Raa'-pspo

RDIjPSP^

RB7/PGD

RBS.;?3C

RS5

R.S4

RB3/PGM

RB2

RBI

RBO/IKT

RD7/PSP7

RDS/PSP8-

RD5/F8PS-

RD4/P3P4

RC7/RX/QT

RC6/TX/CK

RC5/SDO

RC4/SSi''"SDA

RD3/PSF3

RD2/P8P2

Figure 4.13 PIC 16F877Pin Layout

35

Table 4.4 Key Features ofPIC 16F877

Key Features

PiCmicro™ MUMtenge ReferenceManual {DS33023}

PC16F873 PIC16FB74 PIC16F876 PICK3F877

Operating Frequency DC - 20 MHZ DC-20 MHz DC - 20 MHz DC - 20 MHz

RESETS {anrJO slays] FOR. BOR[PWRT. GST)

POR, BOR(PWRT, OST)

POR. BOR

(PWRT, OST;POR. BOR

tPWRT, OST;

FLASH Progran? "emc*v("4-bit words)

4K 4K 8K 8K

'Date Memory (bytes) 192 192 36-5 36S

EEPROM Data Memory 125- 128 256 2ES

.Interrupts 13 '[•4 13 14

I/O ?ort& Ports A=B,C Ports A.B=C,D,£ Ports A.B,C Ports A,BfC3.H

Tmers "; 3 3 3

Ca^ure/ComparefP'ArM Mediates-

2 on

Serai Communications VS3F, JSART MSSP, LJSART MSS= USART MSSP. USA3T

Parallel Communications — PS? —~»o z*

10-bit Analog-to-Dlgltai JVlodu^e 5 input channels 8 mp& channels 5 input channels S input channels

^strjcbcnSet 35 instructions 35 instructions 35 instructions 35

driven by self-developed h-bridge with the combination of TIP 132 and BS 170

transistors.

LMirtjiBm-

FWjtt

na_>u

— n*_f4

H*_*l

rasi

— cees

H»_ffl

•H»_CI

C3

I0UF

hh_b» —

H»JS

«»J»

BUJB

R«UK —

3 ItEl

HliJJr

biijw

HHJH

HILM

fhj»

£-fcfc

C1

U? M«F

TrSm

Figure 4.14 DC Motor Application Schematic Diagram

The movement or rotation of motors is controlled by several inputs, namely

MOTIJUGHT, MOTl_LEFT, MOT2_RIGHT, M0T2JLEFT, MOT3_RIGHT and

MOT3 LEFT. These switches will give high input signal to the input pin of

microcontroller once they are pushed. For example, as the MOTl_RIGHT switch is

pushed, the motor will rotate in clockwise direction and vise versa when

MOTlJLEFT switch is pushed.

The input from switch will give correspondence output to the h-bridge. INI in h-

bridge circuit indicates motor to rotate clockwise. IN2 indicates the motor to rotate

anti-clockwise. As the MOTJRIGHT switch is pushed, it will correspondent^ trigger

high output signal at the outputpin connected to INI.

Thefunction of LMIT_RIGHTandUMITLEFT switch is to stopthe manipulator if

maximum workspace has been achieved. At maximum position for each link,

37

whether rightmost position and leftmost position, the manipulator will hit these limit

switches and automatically the manipulator stops from moving.

Practically, when dc motor is rotating, it will generate pulse from encoder circuit.

This pulse is very useful to determine how much the motor is rotating. The pulse is

then sent to PINA4 and PIN_C0, so that the signal can be processed in such a way

that how much degreethe motor is rotatingcan be determined.

The primary user interface is a RS-232 connection between the MCUand hostPC. A

MAX232 IC is used in the design. The serial port in PC uses the voltages from the

range of-10 V to 10Vwhile IC chips use 0V to 5V. Thus, the ideaof MAX232 IC is

to convertthe voltage signal from -10V- 10Vto 0V - 5V when communicating from

PC to IC chips and vise versa.

The above paragraphs almost describe about the circuit or hardware of controller. For

dc motor software, it performs position calculation and provides a command

interpreter to create and control motion profiles. Dc motor position calculation is

implemented to must perform the following tasks:

• Get current motor position, and

• Get desired motor position.

8. THE PROGRAMMING OF THE CONTROLLER

The general design is the system implements one PIC 16F877 microcontroller. The

microcontroller receives variety of input from control console via various input pins.

Then, the data is processed internally to generate the required output. The function of

microcontroller also to communicate with PC through RS-232 serial communication.

38

8.1 Control Selection

The mode of operation for controller is designed in two states, namely MANUAL

CONTROL and PC CONTROL. During MANUAL CONTROL state, the

manipulator is controlled by using a console. As the manipulator moves, the degree

of rotation is recorded and transmitted via RS-232 Tx pin, so that the manipulator

position is displayed at HyperTerminal or specific user interface. The program only

considers inputs from console and output into Tx pin and ignore whatever incoming

input from PC.

As PC CONTROL is selected, the controlling of manipulator is done totally by PC.

During this condition, all of RS-232 pins are used; Rx, Tx and Interrupt pin. During

this state, the program only considers inputs from PC and output into Tx pin while

ignoring whatever input coming from console. Figure 4.15 shows the flowchart of

Control Selection function.

8.2 Limit Right and Limit Left Function

The limit switch function is designed for several purposes; first, to limit the

movement of the manipulator in its workspace and second, for resetting function of

the manipulator. This section only discusses the limit of the manipulator's movement

while resetting function will be discussed in section 8.3.

From the above section, the workspace is determined as below:

-150° ,

MANUAL CONTROL

Ignores whatever inputfrom PC

Receives input fromconsole

Run the dc motor

according to the inputfrom console

Encoder pulses aregenerated

Calculate the pulsegenerated into equivalent

degree

Transmit to PC throughTx pin for current

position

YES

END

ProgramInitialization

PC CONTROL

Ignores whatever inputfrom console

Receives input from PC

Run the dc motor

according to the inputfrom PC

Encoder pulses aregenerated

Calculate the pulsegenerated into equivalent

degree

Transmit to PC throughTx pin for current

position

YES

END

Figure 4.15 The Control Selection Function Flowchart

40

The idea is every link will be placed two limit switches, placed at both positive and

negative maximum position. When one of those switches is hit, the switch will send

signal to microcontroller to stop the movement. Figure 4.16 shows the flowchart for

limit switch function.

( START J'r

DC motor run

/itch hit?

YES

--___ NO

'

DC motor stop

' r

( END J

Figure 4.16 The Limit Switch Function Flowchart

8.3 Reset Function

The purpose of reset function is to reset all of the manipulator's operation and return

back the manipulator to its original position. The concept is as the reset command is

invoked, the manipulator will move to the rightmost or maximum position until the

maximum limit switch is hit. Then, the dc motor will run on opposite direction until

predefined pulse count is achieved. Now, the manipulator is said to be placed at its

original position. Figure 4.17 below shows the flowchart of reset function for a link.

41

START

Normal operation

YES J*

Manipulatormoves to the rightmostor maximum position

YES

Manipulator moves in the reversedirection from previously

NO

NO

NO

Manipulator stop

END

Figure 4.17 Reset Function Flowchart

8.4 Positioning Function

Positioning function is normal operation for the manipulator. The concept is as user

instructs the manipulator either using console or PC, the program will do as whatuser

wants. Forexample, letsay the user want to move link 1by 30°, the program receives

the instruction and convert this 30° into equivalent pulse count. Then, the

microcontroiier signals the dc motor to rotate. The encoder pulse is generated and fed

into another input pin of microcontroller. If the pulse count is adequate, the motor

42

will stop, indicating that the link 1 has been moved by 30° of rotation. Figure 4.18

shows the flowchart for positioning function.

NO

START

Receive input from PC orconsole

Convert the data into

equivalent pulses

5DC motor run

Motor stop

ITransmit to PC through

Tx pin for currentposition

NO

Figure 4.18 Positioning Function Flowchart

8.5 Emergency Function

The emergency function is purposely for halting immediately the operation of

manipulator. There might be an unexpected occurrence during the operation, such as

data floating or manipulator hitting person in workspace. Figure 4.19 shows the

flowchart for emergency function.

43

START

Positioning Function

YES

Send interrupt tomicrocontroller to stop all dc

motors

END

Figure 4.19 Emergency Function

NO

9. DC MOTOR APPLICATION SOURCE CODE

The code implements a brush dc servomotor using PIC 16F877 microcontroller. The

code also contains the process of controlling dc motor as discussed above.

Initially, the code will instruct the motor to rotate. After that, it receives encoder

pulse signal from the motor for position calculation. Then, the result from position

calculation will be transmitted to PC for displaying purpose through RS-232 serial

communication.

This source code is compiled using PIC C Compiler version 2.7333 available at

Microprocessor Lab. The generated source code is shown in Appendix I

44

10. USER INTERFACE

Instead of console, a HyperTerminal program is used to communicate between PC

and microcontroller. HyperTerminal is used by applying the setting at COMl to

select serial port communication.

Figure 4.20 HyperTerminal Interface, Displaying theCurrent Position for TheManipulator: Link 1 is 30.99° WhileLink 2 is 22.99°.

45

CHAPTER 5

DISCUSSION

The development of robot controller and manipulator involves both electrical and

mechanical skills. In general, the development of controller relates a lot with

electrical skill while development ofmanipulator relates a lot with mechanical skill.

There are several aspects that play important role in determining the successful of

manipulator design, which are stability, reliability, matching and accuracy.

The designed manipulator must be stable in the way that the base can support all three

arms. And the lower arm is able to support correctly the upper arms. This condition

can be achieved by implementing design carefully. The analysis is suggested to be

more done at preliminary and detailed design stage by carefully analyze the full

dimensional drawing. This is because any changes during design stage are easier to

alter and involve lower cost compared in the construction or operation phase. If any

of the designproblems occur during construction phase, the design shouldbe revisited

and do necessary changes. Among the basic design strategies to ensure that the

manipulator is stable are by allocating wide base area and lowering the centre of

gravity at base.

Reliability in manipulator design means the manipulator can easily and properly

operate. If the specified workspace is -90° to +100° (total 190°), the manipulator

must achieve any position in this limit. The disturbance of achieving this limit might

be improper arrangement of parts, for example improper placing of motor that disturb

the movement of arm. This kind of disturbance can be eliminated by designing the

parts' position properly including testing the model and improve the design if any

46

problems arise. The manipulator also should move in such a way that having the

lowest friction at joint and strong link or arm. This consideration can be achieved by

implementing bearing at each joint and strong material is used. Note that, strong

material is important to avoid links' deviation during its operation.

All parts in manipulator must be matched together to maximize its operability. The

parts matching may include all parts are tightly placed at manipulator body,

disallowing vibration between them. In case of vibration is allowed, it may disturb

the rotation or revolution of manipulator. The most important thing is the gear must

be matched in all of its application. These requirements may be achieved by properly

applying screw or bolt and nut to tight the parts together and also use gears with same

pitch only.

The final consideration in manipulator design is it must provide accuracy in

positioning. The accuracy is very difficult to obtain since it involve a lot of criteria.

But the most vital parts to achieve accuracy are the above-mentioned aspects that

have been discussed earlier. Accuracy means how close for the manipulator to

achieve the right position, which is if the controller sends the signal that want to go to

coordinate (3, 4, 5), the manipulator can achieve that point. The most possible

defector for the manipulator to achieve accuracy is the tightening and loosening

process of belt during motor on and off. As the motor on, the belt is tightened while

during motor off, the belt might loose. As the belt loose, the manipulator will slightly

fall down from right coordinate after some manipulator's operation. This problem

should be always reminded by applying belt tightening shaft for all belts used for the

sake of minimizing this effect. This condition is also might be the advantage of using

belt over chain since the belt can be vigorously tightened and chain might little loose

during motor stop.

Another problem that might appear is during power off, which the motor also will off.

Analytically, as the motor off, the motor is not holding arm anymore to a specific

location. Thus, arm will fall down. If the power supply on, the arm then will up

again. In order to remedy this problem, latching gearbox might be used to counter

that effect, which latching gear box will restrict the gear or link to move during power

off.

47

For controller, the determining aspects for the successful of controller are the

goodness of microcontroller programming. The good programming will yield to the

reliability of the program. The program should not mismatch or misinterpret between

user interface (program at PC) and output signal established by microcontroller.

We could say that controller is a heart for manipulator's operation. Whatever things

processed in heart is interpreted in the movement of manipulator. If the program has

very little mismatch between the calculation of desired position and measured

position, this difference will be accumulated throughout the manipulator's movement.

Finally, let say after 50 movements or 60 movements, the error is quite significant.

Thus, it is essential for that program to be equipped together with error calculation

and compensation function.

48

CHAPTER 6

CONCLUSION AND RECOMMENDATION

1. CONCLUSION

As the conclusion, the designing and fabricating process starts from studythe existing

design, followed by preliminary design, detailed design, construction and finally

testing the constructed model.

At study stage, the study of robotic arms or robot manipulators of existing design is

carried out. The functions are to find the idea to develop the designand setting up the

best design parameters.

Then the process is followed by preliminary and detailed design. At this stage, the

drawings and dimensions are established. The design alsoconsiders 3-DOF principle,

selection of robot coordinate frame, workspace determination, forward and inverse

kinematics and gearing design.

After the design is properly established and satisfied, the construction of robot

manipulator is initiated. The construction process is started by selection of best

material that posses light, strong and affordable characteristics. Then, construction

process might be preceded by combining or joining all materials with specific

dimension to be a robot manipulator. During construction stage, these skills are

important such as cutting, drilling, grinding and combining the parts. During

construction stage also, anotheractivities are done such as developing microcontroller

programming, developing the PC interface and generating relevant circuits.

49

The final stage is testing the manipulator. If there are any defects found, the design

should be revisited, altered if necessary and executed on the model.

The controller designing stage is initiated by selection the best MCU in market. Then,

the specification of this MCU must be understood for example its bandwidth, data

memory capacity, program memory capacity and the most important things are pin

layout and registers assignment.

The process of controller design is continued by understanding its program. There are

two ways of developing source code, which are generated ourselves or alteration from

the existing source code. Generating the source code ourselves consumes greater time

than alteration the existing one. There might be nearly same source code in our

resources, e.g book or internet and some alterations to suit our needs must be done.

After the program is generated and compiled with compiler results no error, the

program is ready to be burned in PIC 16FF877. The program is testing again in real

application, with our manipulator and the operation is observed. If the operation does

not meet our design, the program or source code must be reviewed again.

2. RECOMMENDATION

Up to this stage, a robot manipulator and controller is completely designed and

implemented. The application ofrobot manipulator covers only 3-DOF operation. It

is restricted to the positioning function only.

It is recommended that for next stage, this robot manipulator is revamped with

improved design and covers wide application, equipped together with manipulating

function. Thus, the design will require high DOF, 6-DOF or higher. This is

important to observe that better application could be implemented for example the

process of lifting something, drawing, painting and several othersuitable applications.

The design also should be equipped with sensor system, for example it identifies

which material to be picked up and more efficient user interface.

50

REFERENCES

[1] Lung-Wen Tsai, 1999, Robot Analysis: The Mechanics ofSerial and ParallelManipulators, Canada, John Wiley & Sons, Inc., pg 1.

[2] Gordon McComb, 2000, The Robot Builder's Bonanza, Martinsburg,McGraw-Hill, pg. ix.

[3] Saeed B. Niku, 2001, Introduction to Robotics: Analysis, Systems,Applications, New Jersey, Prentice Hall, Inc., pg. 2.

[4] Lung-Wen Tsai, 1999, Robot Analysis: The Mechanics ofSerial and ParallelManipulators, Canada, John Wiley & Sons, Inc., pg2.

[5] Lung-Wen Tsai, 1999, Robot Analysis: The Mechanics ofSerial and ParallelManipulators, Canada, John Wiley & Sons, Inc., pg 46.

[6] Lung-Wen Tsai, 1999, Robot Analysis: The Mechanics ofSerial and ParallelManipulators, Canada, John Wiley &Sons, Inc., pg 25.

[7j Niku,, S., "Scheme for Active Positional Correction of Robot Arms,"Proceedings ofthe 5th International Conference on CAD/CAM, Robotics andFactories ofFuture, Springer Verlag, pg. 590 - 593, 1991.

[8] Brady, M„ J. M. Hollerbach, T. L. Johnson, T. Lozano-Perez, and M. T.

Mason, editors, Robot Motion: Planning and Control, MIT Press, Cambridge,Mass., 1982.

51

Appendix A

Types ofRobot Coordinate Frame

.^*.™\:

ROBOT COORDINATE FRAME

'.'>.-r'-' :,,-'

I

%'.

•':,''.s

Appendix B

Types ofExisting Robot Manipulator in Market

SAMPLE OF CURRENT ROBOTS IN MARKET

Kuka Robot from Kuka Roboter GmbH.

Adept-One Robot from Adept Technology, Inc.

MotomanESI65 from Motoman Inc.

Appendix C

Revolution Joint Indication

Th

eM

an

ipu

lato

r's

An

gle

Assig

nm

en

t

Appendix D

Detailed Drawing ofRobot Manipulator

Fro

nt

Vie

w

ft

=3

L

Leg

end

s)

A-

RS

dc

mo

tor

wit

had

d-o

nen

co

der

B-

4N

nad

ap

tab

leg

earb

ox

C-

Co

ntr

oll

er

bo

xD

-B

elt

an

dg

ear

E-

Mo

tor

mo

un

tin

gp

late

No

te"

All

dim

en

sio

ns

are

(nin

ch

Sid

eV

iew

2.4

00

0

«3

00

0

Leg

en

dsi

A-

RS

dc

no

tor

wit

had

d-o

nen

co

der

B-

4N

nad

ap

tab

leg

earb

ox

C-

Co

ntr

oll

er

bo

xD

-B

elt

an

dg

ear

EI-

Mo

tor

mo

un

tin

gp

late

No

tei

AU

dim

en

sio

ns

are

InIn

ch

OO

99

r-0

1.3

80

0

11

.00

00

•9.0

00

0-

op

vie

w

•15

.67

50

- L ZEL

eg

en

ds:

A-

RS

dc

no

tor

wit

had

d-o

nen

co

der

B-

4N

nad

ap

talo

leg

earb

ox

C-

Co

ntr

oll

er

bo

xD

-B

elt

an

dg

ear

E-

Mo

tor

mo

un

tin

gp

late

No

te:

Atl

dim

en

sio

ns

are

inin

ch

3.1

50

0

Appendix E

Robot Manipulator's Workspace

To

pV

iew

Wo

rksp

ace

Appendix F

Detailed Calculation ofPosition Analysis

DETAILED CALCULATION OF POSITION ANALYSIS

10

i

r'(7,4,10)

x

y

*

/74

1

r

Analyzingx- and z-axis

Z A

10(7, 4, 10)

Analyzingx- andy-axis

y

Setting upa parameter:LengthA-B =4.7

-l#2=90°-cos

= 16.71°

J238

Appendix G

Detailed Drawing ofFirst Manipulator Design

150

1,4748

50

1.7150

(_

•3.4380-

_>D _

—0,3750

•7.0000

H h-0.3580— 4.6650-

•10.0000-

B

Robot Manipulator: Front View

LegendsiA - Controller BoxB - Gear

C - Pittman Lo-Cog MotorD - RS Adaptable Gearbox€ - Copper RodF - Koyo Bearing & Bearing PlasticHolder

G - Belt Tightening RodH - Belt

Material used* AluminiumNotei All dimensions are in inch

3.3150

0.7150 -1

5.0000

1.6350

8.000

5.0000

•4.0000-

-M7190

Robot Manipulator' Side View

Legends)BoxA - Controller

B - GearC - Plttnan Lo-Cog MotorD - RS Adaptable GearboxE - Copper RoadF - Koyo Bearing 8> BearingPlastic HolderG - Belt Tightening RoadH - Belt

Material used" Alumlnlun

Notet AU dlnenslons areinch

B.0000

In

B m

00.7158.

^i

ik —'A01.6250

i

tf-w-r-j

•L6150-J

-6.0000-

-8.0000-

5.0000

4.0000

10

.00

00

3C

B

04

.66

50

-

-8.0

00

0-

I1

-E

2=

TK

?-6

.00

00

-

1

0.^5

93

Ro

bo

tM

an

ipu

lato

r!T

op

Vie

w

Leg

en

ds'

A-

Co

ntr

oll

er

Bo

xB

-G

ea

rs

C-

Plt

tnan

Lo

-Co

gM

oto

rD

-R

SA

da

pta

ble

Gea

rbo

xE

-C

op

per

Ro

dF

-K

oyo

Bea

rin

g8.

Bea

rin

gP

last

icH

old

erG

-B

elt

Tig

hte

nin

gR

od

H-

Belt

Ma

teria

lu

sed

'A

lum

inu

m

No

te'

All

dim

en

sio

ns

are

InIn

ch

F

-5.0

00

0-

-8.0

00

0-

Appendix H

DC Motor Application Schematic Diagram

LIM

IT1_

RIG

HT

-O

O

LIM

IT1_

LE

FT

-OO

-

M0

T1

_R

IGH

T

-OO

M0

T1

_L

EF

T

OO

-

Osc

4M

HZ

LIM

IT2_

RIG

HT

-OO

-

LIM

IT2_

LE

FT -O

O

M0

T2

RIG

HT -O

O-

MO

T2_

LE

FT -

OO

-

L1M

IT3_

RIG

HT

-OO

LIM

IT3_

LE

FT -o

o-

Pu

lse

INE

NC

0D

ER

1

Pu

tse

INE

NC

OD

ER

2

mo

t:i_

RIG

HT

MC

LR

1

PIN

_A0

PIN

_A1

PlN

J\2

PIN

A3

PtN

_A

4

•PIN

JUS

PIN

J30

PIN

_E1

PlN

_E2

Vd

d

.Vss

0S

C1

OS

C2

PIN

-CO

PIN

,C1

PIN

C2

PtN

_C3

WN

_D

0

-PIN

D1

oo

M0

T3

_L

EF

T

P1

C1

6F

87

7

PIN

.B7

PIN

J36

PIN

_8B

PIN

_B4

PIN

_B3

P1N

_B2

PIN

_B1

PIN

_B0

Vd

d

Vss

PIN

_D7

PIN

_D6

PIN

_D6

PIN

_D4

PIN

_C7

PIN

_C6

PIN

_C5

PIN

„C4

PIN

_D3

PIN

D2

fr 1

0U

F

10

uF

-O

o_

C3

H-

!

10

uF

C1

U7

10uF

C1

+

V-

C1

-

-C2

+

C2

-

-V-

T2

ou

t

R2

in

MA

23

2

Vcc

Vss

T1

ou

t

R1

in

R1

ou

t

Tli

n

T2

in

R2

oti

t

vcc

SV

J9

_ o

o o--

o

DS

UB

9F

INI

IN2

1N1

IN2

IN1

IN2

12

V

Vcc

0U

T2

OU

T1

OU

T2

OU

T1

BR

IDG

E1

12

V

Vcc

OU

T2

0U

T1

HB

RID

GE

2

H_B

RID

GE

3

Til

le:

DC

Mo

App

lica

tion

-R

ob

otC

on

tro

ller

&M

anip

ulat

or

UT

PF

inal

Yea

rP

roje

ct

Des

ign

edby

:Z

ulki

fli

Ch

eck

ed

by;

App

rove

dby

:

Do

cu

men

tN

00

01

Date

Jun

09

,2

00

4

Sh

eet

1o

f1

Rev

isio

n1

.0

Siz

eA

Appendix I

Program Source Code

ftinclude

#fuses XT, NOPROTECT, NOWDT#use delay(clock=4000000)#use rs232(baud=9600, xmit=PIN_C6, rcv=PIN_C7

float movement_l();float movement_2 ();void reset {) ;

//declaration of motor 1#define LIMIT1JUGHT PIN^AO#define LIMITl_LEFT PIN_A1#define M0T1_RIGHT PIN_A2#define M0T1_LEFT PIN_A3#define IN_ENC0DER1 PIN_A4#define H_BRIDGE1_1 PIN__B4#define H_BRIDGE1_2 PIN_B5

//declaration of motor 2#define LIMIT2_RIGHT PIN__C1#define LIMIT2_LEFT PIN_C2#define MOT2_RIGHT PIN_C3#define MOT2_LEFT PIN_C4#define IN_ENCODBR2 PIN_C0#define H_BRIDGE2_1 PIN_D4#define H_BRIDGE2_2 PIN_D5

//declaration of motor 3#define MOT3_RIGHT PIN_D2#define MOT3_LEFT PIN_C5#define H_BRIDGE3_1 PIN_D6#define H_BRIDGE3_2 PIN_D7

//general declaration#define IN RESET PIN A5

//main function

main()

{float position;

set_tris_a(Oxff) ;set_trisjD(OxOO);set_tris_c(Oxff);set_tris_d(OxfO);

printf("\nRobot Controller and Manipulator\n");printf("\nCalculating the degree of rotation of robot manipulator\n");

// controlling motor 1while (true)

{if (MOTl_RIGHT)

{if (!LIMIT1_RIGHT && !LIMITl_LEFT)

output_high(H_BRIDGElJL);

if (LIMIT1_RIGHT)output_low(H_BRIDGEl_l);

if (LIMIT1_LEFT)output_high{H_BRIDGEl_l);

}else

output_low(H_BRIDGEl_l);

if (MOTl_LEFT){if (!LIMIT1__RIGHT && !LIMIT1_LEFT)

output_high{H_BRIDGEl_2);

if (LIMIT1_LEFT)output_low(H_BRIDGE1_2);

if (LIMIT1_RIGHT)output high(H_BRIDGEI_2);

}else

output_low(H_BRIDGEl_2);

//controlling motor 2if (MOT2__RIGHT)

{if (!LIMIT2_RIGHT && !LIMIT2_LEFT)

output_high(H_BRIDGE2_l);

if (LIMIT2_RIGHT)output_low(H__BRIDGE2_1);

if (LIMIT2_LEFT)output_high(H_BRIDGE2_l);

}else

output_low(H_BRIDGE2_1);

if (MOT2_LEFT){if (!LIMIT2_RIGHT && !LIMIT2_LEFT)

outputJiigh(H_BRIDGE2_2);

if (LIMIT2_LEFT)output_low(H_BRIDGE2_2);

if (LIMIT2_RIGHT)output_high(H_BRIDGE2_2) ;

}else

output_low(H_BRIDGE2_2) ;

//controlling motor 3if (MOT3_RIGHT)

outputJligh(H__BRIDGE3_l) ;

if (MOT3_LEFT)output_high(H_BRIDGE3_2);

position_l=movement_l();position_2=movement_2();

printf("\n The current position for link 1 is %f degrees.\n", positional);printf("\n The current position for link 2 is %f degrees.\n", position_2);

reset() ;

}

>

//Calculating the degree of rotation-

float movement_l(){float pos, count_pulse;

pos=0, count_pulse=0;

while (input(MOTl_RIGHT) ){

if (input(IN_ENC0DER1))count_pulse = count_pulse + 1;

}

while (input(M0T1_LEFT)J{

if (input(IN_ENC0DER1))count_pulse = count_pulse - 1;

}

}

pos = count_pulse/2; //2 pulses = 1 degree of rotationreturn pos;

float movement_2()

{float pos, count pulse;

pos=0, count_pulse=0;

while (input(M0T2_RIGHT)){

if (input(IN_ENC0DER2))count__pulse = count_pulse

}

while (input(M0T2_LEFT))

{if (input(IN_ENC0DER2))count_pulse = count_pulse

}

pos = count_pulse/2;return pos;

//Function for resetting-

void reset(void)

{int i, j;while (IN_RESET)

{if (!input(LIMITl^RIGHT))

output_high(H_BRIDGE1_1) ;

for (i=300;i==0;i—)

output_high(H_BRIDGEl_2);

if (!input(LIMIT2_RIGHT))output high(H BRIDGE2 1);

for (j=180;j==0; j—)output_high(H_BRIDGE2_2);

}

l;

- 1;

//2 pulses = 1 degree of rotation

//calculating the rotation pulse to initial state